dg :: software :: aos :: 093-000195-02 AOS Debugger and Disk File Editor UM Jun78 raw

Advanced

Operating System

(AOS)

Debugger

and

Disk File Editor

User's Manual

093-000195-02

For the latest enhancements, cautions, documentation
changes, and other information on this product, please see
the Release Notice (08 5 -series) supplied with the software.

Ordering No. 093-000195

©Data General Corporation, 1976 1977, 1978

All Rights Reserved

Printed in the United States of America

Revision 02, June 1978

Licensed Material - Property of Data General Corporation

DataGeneral

software documentation Licensed Material - Property of Data General Corporation

NOTICE

Data General Corporation (DGC) has prepared this manual for use by DGC personnel, licensees,
and customers. The information contained herein is the property of DGC and shall not be
reproduced in whole or in part without DGC prior written approval.

DGC reserves the right to make changes without notice in the specifications and materials
contained herein and shall not be responsible for any damages (including consequential) caused by
reliance on the materials presented, including but not limited to typographical, arithmetic, or listing
errors.

Advanced Operating System

(AOS)

Debugger and Disk File Editor

User's Manual

093-000195

Revision History:

Original Release - April 1976
First Revision - April 1977
Second Revision - June 1978

This document has been extensively revised from revision 01; therefore,
change indicators have not been used.

The following are trademarks of Data General Corporation, Westboro, Massachusetts:

U.S. Registered Trademarks Trademarks

CONTOUR I INFOS NOVALITE DASHER

DATAPREP NOVA SUPERNOVA microNOVA

ECLIPSE NOVADISC

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Preface

This manual describes the Advanced Operating System
(AOS) Debugger and its companion utility, the AOS
Disk File Editor (called DEDIT). Both are interactive
programs which you execute from the Command Line
Interpreter (CLI). The Disk File Editor commands are
a subset of Debugger commands. In AOS, executable
programs (.PR files) are stored on disk precisely as they
will be brought into memory and executed.
Corresponding addresses on disk and in main memory
have the same contents. Commands that affect memory
locations in the Debugger affect disk locations when
you execute DEDIT. The changes you make with the
Debugger are effective only for that run of the
Debugger; when you exit from it, the changes are
wiped out. DEDIT changes, however, and Debugger
changes in a shared area, are permanent and remain in
the disk file.

You can use DEDIT to edit any kind of file on disk (a
data file, for example). Normally, you correct a program
using the Debugger, test it, and then make the changes
permanent; you can make permanent changes to the
disk file using DEDIT. Eventually, you will want to
make permanent changes to the source program using
the text editor; you then reassemble or recompile the
source program and rebind it.

The Debugger and Disk File Editor User's Manual is
organized as follows:

Chapter 1 Introduces the Debugger utility. All
material in Chapter 1, except for the
command that calls the Debugger and the
material on breakpoints and on starting
user program execution, applies also to
the Disk File Editor.

Chapter 2 Describes arithmetic, logical and Boolean
expressions. You can use these in both
DEBUG and DEDIT.

Chapter 3 Explains the DEBUG/DEDIT 2-part
address mechanism which allows you to
access any bit, byte or word location in
your memory area (Debugger) or disk file
(DEDIT).

Chapter 4 Describes Debugger commands. Those
that do not apply to the Disk File Editor
are noted.

Chapter 5 Explains DEDIT -- how to invoke DEDIT
from the CLI.

Although you can debug programs written in
higher-level languages, like FORTRAN, you must
know what assembly-language statements your
compiler produces. To correct higher-level language
programs, you may find that the runtime error
messages guide you to problem areas more quickly.
Your compiler manual will help you interpret these
messages.

If you are completely unfamiliar with the debugging
process, read the chapter in Learning to Use Your
Advanced Operating System (093-000196) which contains
a sample debugging session on an assembly-language
program.

Appendix A Explains the DEBUG and DEDIT error
messages.

Appendix B Summarizes DEBUG and DEDIT
commands for easy reference.

Appendix C Explains how to check the ANSI status of
your console, since it determines which
key you press for the new-line and
(CR/LF) functions.

093-000195-02

Preface

DataGeneral

SOFTWARE DOCUMENTATION

Correcting Typing Errors

You can correct your typing errors in either of the
following ways:

• Press the RUBOUT key to delete the
previously-typed character and backspace the cursor.
For example:

P 302 (RUBOUT) (RUBOUT))

deletes 2 and and repositions the cursor:

P3_

• Press CTRL-U to delete the entire command line.

Reader, Please Note:

The Debugger and Disk File Editor differ from other
AOS programs (like the CLI) in that they do not treat
the new-line, carriage-return and line feed keys on your
console the same.

In this manual, the key you press to type an ASCII 012
is called a new-line and is represented in formats as }.
The key you press to type an ASCII 015 is represented
in this book as (CR/LF). You use a new-line to delimit
most DEBUG/DEDIT commands; (CR/LF) is itself a
command.

ANSI-standard terminals have a new-line key and a
carriage-return key on their keyboards; non-ANSI
terminals have a carriage-return key and a line-feed
key. Appendix B describes how to tell whether your
terminal is an ANSI or non-ANSI standard model and
whether its device characteristics are properly matched;
it describes which key you would press for a new-line
character and which for a (CR/LF).

Licensed Material - Property of Data General Corporation

Other notation conventions we use in this manual are:

COMMAND required [optional] ...

Where Means

COMMAND You must enter the command (or
its accepted abbreviation) as
shown.

required You must enter some argument

(such as a Filename). Sometimes,
we use:

required 1
required 2

which means you must enter one of
the arguments. Don't enter the
braces; they only set off the choice.

[optional] You have the option of entering

some argument. Don't enter the
brackets; they only set off what's
optional.

You may repeat the preceding
entry or entries. The explanation
will tell you exactly what you may
repeat.

I SHIFT-N. Press the SHIFT and N

keys to produce j.

All numbers are octal unless we indicate otherwise by
using a decimal point; 3., for example.

Finally, we usually show all examples of entries and
system responses in THIS TYPEFACE. But, where we
must clearly differentiate your entries from system
responses in a dialog, we will use

THIS TYPEFACE TO SHOW YOUR ENTRY)

THIS TYPEFACE FOR THE SYSTEM RESPONSE

093-000195-02

DataGeneral

Licensed Material - Property of Data General Corporation software documentation

Contents

Chapter 1 - Introduction

Entering and Terminating the Debugger l_l

Debugger Command Formats \.\

Temporary Variables l_2

Debugger Error Response 1_2

Chapter 2 - Expressions

Definition of an Expression 2-1

Evaluating an Expression 2-1

Definition of Expression Operators 2-1

Arithmetic Operators 2-1

Indirect Operator @ 2-2

Logical Operators 2-2

Boolean Arithmetic Operators 2-3

Hierarchy of Operator Evaluation 2-3

Using Assembler Instructions in Expressions 2-4

Characters 2-4

Chapter 3 - Debugger Addressing

DOT Symbol (Period) 3 . 2

Chapter 4 - Debugger Commands

Breakpoint-Related Commands 4_1

Set a Breakpoint 4_1

Examine Breakpoints 4_2

Delete Breakpoints 4_3

Start User Program Execution 4.3

Set Variable Command 4.3

Display Accumulator Command 4_3

093-000195-02

Contents

DataGeneral

software documentation Licensed Material - Property of Data General Corporation

Chapter 4 (continued)

Location-Related Commands 4-3

Display Contents of a Location 4-3

Modify Contents of a Location 4-4

Display the Next Data Item 4-4

Display the Previous Data Item 4-4

Display a Range of Data Items 4-5

Suppress Symbols 4-5

Display an ASCII String 4-6

Mode-Related Commands 4-6

Display Current Display Modes 4-7

Display Last Item with Different Display Modes 4-7

Floating-Point Accumulator-Related Commands 4-8

Set Floating-Point Accumulator 4-8

Display Contents of a Floating-Point Accumulator 4-8

Floating-Point Status 4-8

Display Linked Elements 4-8

Compute an Expression and Display its Result 4-9

Interpret an Error Code 4-9

Append a Symbol Table 4-9

LOG Commands 4-9

Debugging Shared Libraries 4-9

Terminate the Debugger 4-10

Chapter 5 - AOS Disk File Editor (DEDIT)

Appendix A - Error Messages

Appendix B - DEBUG/DEDIT Command Formats

Breakpoint Commands (Do not apply to DEDIT) B-l

All Other Commands B-l

Appendix C - Command Terminator Keys and ANSI and
Non-ANSI Standard Consoles

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Chapter 1
Introduction

Debugging is the process of detecting, locating, and
removing errors from your program. When you are
debugging a program, you can control its execution and
halt it at specified addresses.

With the debugging commands, you can:

• Monitor and alter memory locations, accumulators,
and the carry bit.

• Set, delete, and examine breakpoints.

• Restart program execution at desired points within
the program.

• Set the Debugger output format.

• Display portions
debugging it.

After you determine the addresses and set the desired
breakpoints, you start execution of the program. When
it reaches a breakpoint, the program halts before
executing the instruction at the breakpoint location. At
this point you issue debugging commands. Then you
can restart program execution either at this address, or
at some other address within the program.

Entering and Terminating the
Debugger

To debug a program, enter the following CLI
command:

DEBUG pathname [arguments-for-process] )

pathname is either a filename in the working directory,
or a pathname indicating the location of the file outside
of the working directory. Pathnames are fully explained
in the A OS Programmer's Manual.

For example:

DEBUG MYPROGRAM)

loads MYPROGRAM and enters the Debugger.

You can also enter the Debugger by coding the system
call ? DEBUG in your program at the point where you
want to transfer control to the Debugger. Like other
system calls, 7DEBUG has an error and normal return,
and when you key the P (proceed) Debugger
command, the Debugger will return control to your
program at the normal return.

After you successfully enter the Debugger, it displays
the following prompt:

of the program while you're +

You then key in debugging commands immediately
following the plus sign.

When you have completed debugging your program,
return to the CLI by keying the command.

BYE)

Debugger Command Formats

one of two

You input a Debugger command
different formats:

in

Type 1. command-code [u arg [;arg]...] )
Type 2. [arg[;argj... J keystroke-command

A Type 1 command consists of a command-code,
followed optionally by a blank and a list of from one to
n arguments. When you supply two or more
arguments, you must separate them with semicolons
(;). You terminate the command line by keying in a
new-line character ( ) ) as described in the Preface.

093-000195-02

1-1

Debugger Command Formats

DataGeneral

SOFTWARE DOCUMENTATION

For example:

P)

The P command-code directs the Debugger to proceed
with program execution. In this example the program
begins program execution at the address contained in
the location counter.

A Type 2 command consists of an optional argument
list (where optional arguments are in lowercase italic
letters), followed by a single-character
keystroke-command. An argument consists of one or
more ASCII characters. When you supply two or more
arguments, you must separate them with semicolons
(;). The single keystroke-command determines the
action. The boldface lowercase words in the format
indicate that you must supply both the appropriate
arguments and the keystroke-command terminator. In
this manual, user input IS IN THIS TYPEFACE;
Debugger output IS IN THIS TYPEFACE.

For example:

3105:000075 +

The Debugger displays the contents of the location
referenced by octal address 3105; the colon (:)
keystroke-command specifies the display action. The
contents of the location are displayed after you type the
colon.

You can also use the colon alone (without an
argument). In this case, the Debugger uses the
contents of the last displayed location as the address of
the next location to be displayed.

Example:

3105: 000050 + (display contents of location 3105)
: 000007 (display contents of location 50)

Other keystroke-commands include the equal sign
( = ), (CR/LF), and | (SHIFT-N).

Temporary Variables

When you use the Debugger, you can
temporary variables using the SET command.

define

Licensed Material - Property of Data General Corporation

command. The values of any temporary variables are
maintained only until you return to the CLI.

The following special symbols are the temporary
variables maintained by the Debugger. You can use
them in an expression as you would any symbol. (Note:
You cannot use these special Debugger variables in the
Disk File Editor, but you can still use the SET
command to define your own temporary variables
when you are in DEDIT.)

Symbol Represents

#P

Current program counter (points to a
breakpoint when the program halts)

#C
#0

#1

#2
#3
#R

The Debugger itself maintains a special set of
temporary variables. You can change these variables,
just as you would your own variables, using the SET

Carry bit

Accumulator (AC0)

Accumulator 1 (AC1)

Accumulator 2 (AC2)

Accumulator 3 (AC3)

Result of the last-executed " = " command

If you have the floating-point option:

Symbol Represents

#FS1 High-order part of floating-point status
#FS2 Low-order part of floating-point status

Examples:

#1 = 001400

Displays the contents of accumulator 1.

#1%EQ%1400 = 000001

In this example the contents of accumulator 1 are
001400. The comparison is true.

Debugger Error Response

The Debugger checks all command lines for syntax
errors. If it detects an error, it displays an appropriate
error message; you can then enter the correct
command. For a list of Debugger error messages and
their meanings, see Appendix A.

End of Chapter

1-2

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Chapter 2
Expressions

Definition of an Expression

An expression is a combination of variables, constants,
special Debugger symbols, user-program symbols, and
operators. Spaces can be used in expressions for
legibility and are optional items.

Evaluating an Expression

You can have the Debugger evaluate an expression and
either display the result, or use the result as an address
and display the contents of that location. In both cases,
the format is:

expressions

> RETURNED VALUE

Using the equal sign will give you the numerical value
of the expression, while the colon will give you the
contents of the location symbolized by the expression.

Example:

Given the following locations and their contents

Address

Contents

0015

000035

0016

002002

0017

010236

0020

000007

15 + 3= 20

15 + 3: 000007 +

The octal digits 1 5 and 3 are the expression constants,
the plus sign ( + ) is the arithmetic operator, and the
equal sign and colon are the keystroke-commands.
Notice that you do not use the new-line key ( ) ) after
the keystroke command.

NOTE: The Debugger accepts constants as octal values.
You can, however, enter decimal constants by
keying one or more digits, immediately
followed by a decimal point.

Definition of Expression Operators

The Debugger recognizes four types of expression
operators:

• Arithmetic

• Indirect operator ( (i

• Logical

• Boolean arithmetic

Operator

+

/
()

Arithmetic Operators

Operation

Addition and unary plus

Subtraction and unary minus

Multiplication

Division

Parentheses, when used algebraically.

NOTE: If you use the division operator (/), the type of
divide performed depends on the sign submode
(see "Changing Display Modes" in Chapter 4).
If the sign submode is US, an unsigned divide is
performed. If the submode is SI, a signed divide
is performed.

Examples:

3 + 2 + 1= 6

Octal constants and octal result.

10-3= 5

Octal constants and octal result.

8. + 2.= 12

Decimal constants and octal result.

093-000195-02

2-1

Arithmetic Operators

DataGeneral

SOFTWARE DOCUMENTATION
11/3= 3

Octal constants and octal result.

7/2 = 3

Octal constants and octal result. The Debugger
truncates the result.

2*3 + 4= 12

Octal constants and octal result.

2*(3 + 4)= 16

Octal constants and octal result.

Assume the following ,is part of an assembly-language
program:

.ENTABC
.NREL

ABC: LDAO,B
LDA1,C
ADD 1,0
MOV 0,2

B:
C:

3
4

You can use the symbol ABC in an expression since the
symbol is referenced in the .ENT statement. Assume
the instruction LDA 0,B has been loaded in the user
process area at location 1400. Then

ABC= 1400

displays the address of ABC, and

ABC + 3 = 1403

displays the address of the MOV 0,2 instruction.

Indirect Operator @

When you use the commercial AT sign ( @ ) within an
expression, the Debugger evaluates the expression
using the contents of a location. The AT sign must
immediately precede the location's address, and
implies "use the contents of the location instead of the
address".

Licensed Material - Property of Data General Corporation

Examples:

Given the following addresses and their contents:

Address Contents

001400

001500

001401

000007

001 500

001600

001 600

000050

@1400 + 1 =

= 1501

Add 1 to the contents of location 1400
(1500+1 = 1501).

@(1400+1)= 7

Note that the Debugger does not evaluate this in the
same way as the first example, but evaluates the
expression in the following manner: first it evaluates
the parenthetical expression (1400+1) and uses this as
the address, then it displays the contents of location
1401.

@ 1400: 001600 +

uses contents of location 001400 (1500) as the address
and displays contents of location 001500.

@@1 400: 000050 +

This illustrates a double level of indirection. The
Debugger uses the contents of location 1400 (1500) as
the first address; uses the contents of that location
(1600) as the second address; and displays the contents
of that location.

Logical Operators

Logical operators compare values. When you use a
logical operator in an expression, the comparison
returns a value of 1 if true, and if false. The Debugger
compares expression values following algebraic rules.

The following logical operators compare two
signed integers (in twos-complement form).

Operator Operation

16-bit

%EQ%

Equal to

%LT%

Less than

%LE%

Less than or equal to

%GE%

Greater than or equal to

%GT%

Greater than

%NE%

Not equal

2-2

093-000195-02

Licensed Material - Property of Data General Corporation

Examples:

5%EQ%5= 1

5%GT%6=

6%NE%5= 1

0%GT%-1 = 1

Note that you enter a negative number by preceding
the value with a minus sign (-).

@1401%EQ%7= 1

In this example the contents of location 1401 are
000007. The Debugger evaluates the expression using
the location's contents.

The following logical operators compare two 16-bit
unsigned integers.

Operator Operation

%EQ% Equal to

%ULT% Less than

%ULE% Less than or equal to

%UGE% Greater than or equal to

%UGT% Greater than

%NE% Not equal

Note that the EQ and NE operators are the same for
both signed and unsigned logical compare operations.

Examples:

5%EQ%6=

6%NE%7= 1

5%UGE%5= 1

3%UGT%6=

8.%ULE%11= 1

Decimal 8 less than octal 11

-1%UGT%0= 1

This and the example below are true because the
comparison is of 16-bit integers, with no sign bit.

-3%ULE%5= 1

DataGeneral

SOFTWARE DOCUMENTATION

Boolean Arithmetic Operators

The following operators perform Boolean arithmetic
operations.

Operator Operation

%AND% Logical and
%OR% Inclusive or
%XOR% Exclusive or

Examples:

6%AND%3= 2

14%OR%3= 17

13%OR%6= 17

12%XOR%6= 14

(5%LT%6)%OR%(3%GE%7)= 1

The expression is evaluated as follows:

(5%LT%6)%OR%(3%GE%7)= 1

1 %OR%

1 = 1

(7%GT%1 0)%AND%(6%ULE%4) =

-0

Hierarchy of Operator Evaluation

The Debugger evaluates expression operators in the
sequence shown in the following table. You can enclose
portions in parentheses (as in algebraic notation) to
modify the order of evaluation.

Operator

Operation

Indirect

+,-

Unary plus and minus

V

Multiplication and division

+,-

Addition and subtraction

%EQ%,%LT%,%LE%,
%GE%,%GT%,%NE%,
%ULT%,%ULE%,
%UGE%,%UGT%

Arithmetic comparisons (all
of equal priority)

%AND%, %OR%, %XOR% Boolean

093-000195-02

2-3

Hierarchy of Operator Evaluation

DataGeneral

SOFTWARE DOCUMENTATION

Examples:

5 + 3*2= 13

5 + 3*-2 = 177777

6+3%AND%5+5= 10

5%LT%6%AND%79.%ULE%4%OR%10%NE%7= 1

This expression is evaluated as follows:

5%LT%6%AND%79.%ULE%4%OR%10%NE%7= 1
1 %AND% %OR% 1
%OR% 1

1 = 1

Any intermediate values the Debugger generates while
computing an expression's value will be truncated to
the least significant 16 bits.

Using Assembler Instructions in
Expressions

You can use Macroassembler instructions in Debugger
expressions. To do so, you must enclose the instruction
within angle brackets, < >. The Debugger then
assembles the instruction and evaluates it as a 16-bit
integer, which it uses in further computations in place
of the instruction. The symbols # and @, when used
inside angle brackets, retain the meaning assigned to
them by the Macroassembler.

Licensed Material - Property of Data General Corporation

Examples:

<LDA3 16>+5= 34023

<JSR @ 17>= 6017

Characters

You can type any ASCII characters which are not
self-delimiting commands (comma, colon, etc.). To
insert a single ASCII character, precede it with the
quote character ("). To insert a pair of ASCII characters
in a word, use the format a*400 + b where a and b are
the characters to be inserted. You can use characters in
this form in any expression.

The following example displays location 1000 and
replaces its contents with ASCII " A":

1000: 000000+ "A)

In the example below, the Debugger again displays
location 1000 and replaces its contents with ASCII
"A". This time the (CR/LF) delimiter displays the
next location's contents.

1000: 000000+ "A(CR/LF)
1001:000401 +

In the following example, the Debugger inserts an
ASCII "A" in the left byte of location 1000 and a "B"
in the right byte.

4- 1000: 000000+ "A*400 + "B)

The CR/LF delimiter could be used instead of new-line
). In this case, the Debugger would also display the
next location's contents.

End of Chapter

2-4

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Chapter 3
Debugger Addressing

The Debugger uses a two-part address to reference
locations in the user process area. The first part
(base-address) is a number or symbol that references a
16-bit word; the second part (offset) is the number of
words, bytes, or bits added to the base-address. By
using this two-part address, you can reference any bit,
byte, or word location in the user process area.

"Address" in this manual indicates you can use any of
the following addressing formats (where base-address
and offset refer to any valid Debugger expression):

base-address &address-mode-character offset

base-address &address-mode-character

&address-mode-character offset

offset (Uses preset default address mode.)

The base-address is the address of a word in the user
process area. The ampersand (&) indicates that the
following character is an address-mode character, and
offset is the number of words, bytes, or bits added to
the base-address (as specified by the address-mode
character) .

The address-mode character specifies the type of offset
you want to enter (words, bits, or bytes). You can use
the address-mode character with both base-address and
offset, or with the base-address alone, or with the
offset alone. When you enter the Debugger, the
default address mode references a 16-bit word. You can
change this default mode with the MODE command
described in Chapter 4. When you use the
address-mode character in an address, it temporarily
overrides the previously-set default address mode, but
is active only for the expression you are entering.

The address-mode characters are:

Character Action

W Offset references a 16-bit word.

Y Offset references an 8-bit byte.

I Offset references a single bit.

Examples:

Given the following locations in the user process area:

Address Contents

001400 001500

001401 000007

001402 005001

001403 001600

001404 000050

1 400 &W 1:000007 +

adds 1 to base-address 1 400 and displays its contents.

1402&W-1: 000007 +

adds -1 to base-address 1 402 and displays its contents.

Note that in both of the above examples, the offset
added to the base-address is in words (&W).

1401 &Y7: 050 +

adds 7 bytes (&Y) to base-address 1 401 and displays the
referenced byte (second byte in 1404).

1400&I6: 7 +

adds 6 bits to base-address 1400 and displays the
referenced bit (bit 6 in location 1400).

093-000195-02

3-1

Debugger Addressing

DataGeneral

SOFTWARE DOCUMENTATION

1 400 &W: 001500 +

displays the contents of location 1 400. When you omit
the offset from the address, the offset is assumed to be
0.

1400&Y: 003 +

displays byte (left byte) of location 1 400.

&Y3003: 007 +

displays right byte of word 1401 (byte 3003 in your
address space). When you omit the base-address, it is
assumed to be 0.

&Y 1401*2 + 1: 007 +

also displays right byte of word 1 401 .

&l 30006: 1 +

displays bit 6 of word 1400 (bit 30006 in your address
space).

1400:007500 +

In this example the address-mode character is omitted,
so the Debugger used the previously-set address mode
(in this case, word addressing).

Licensed Material - Property of Data General Corporation

DOT Symbol (Period)

The DOT symbol (.) is the address of the last location
referenced by a Display or Modify command (see
Chapter 4), and you can use it in any valid Debugger
expression. If you use it as part of an address that
contains an address-mode character, or in an arithmetic
expression, then the Debugger uses only the
base-address portion of the address of the
last-referenced location.

Examples:

1400: 001500 +
..001500 +

. + 2:005001 +

.&Y5:050 +

1400:007500 +
@. + 3= 1503

End of Chapter

3-2

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Chapter 4
Debugger Commands

Breakpoint-Related Commands

When an executing program encounters a breakpoint,
it immediately passes control to the Debugger before
executing the instruction at the breakpoint location.
You can then key in debugging commands to examine
and modify memory locations, accumulators, the carry
bit, and the program counter. Then you can restart the
program at the breakpoint, or at any other location
within the program.

Set a Breakpoint

You can set as many breakpoints as you need, and you
can specify if a breakpoint is conditional. For a
conditional breakpoint, you supply an expression as
part of your breakpoint-setting command. When the
breakpoint is reached, the Debugger evaluates the
expression and either skips or stops at the breakpoint
according to the result of the evaluation. Figure 4-1 is a
diagram of the logic for breakpoints set with Debugger.

To set a breakpoint, key in a breakpoint command in
the following format:

B [address] [;breakpoint-condition][;breakpoint-count] }

where breakpoint-condition and breakpoint-count can be
any valid Debugger expression, and address must be
the address of a 16-bit word.

The Debugger will:

1. Place the breakpoint at the location specified by
address. If address is omitted, the Debugger will
place the breakpoint at the last-referenced word.

2. Analyze and save the breakpoint-condition as a 16-bit
unsigned integer. This tells the Debugger the
circumstances under which the breakpoint will be
taken. If you leave out breakpoint-condition, the
breakpoint is unconditional.

3. Analyze and save the breakpoint-count. This is the
number of times that the Debugger will bypass the
breakpoint (when the breakpoint-condition is true)
before taking it. If there is no breakpoint-count, the
Debugger automatically sets the count to 1.

4. When your executing program encounters the
breakpoint, the Debugger evaluates the
breakpoint-condition in terms of the existing
program state (accumulators, program counter,
contents of location, etc.). If the evaluation of the
breakpoint-condition produces a zero as the result,
the Debugger will skip the breakpoint and proceed
with the program. Only if the result is nonzero
(true) will the Debugger decrement the
breakpoint-count. When the breakpoint-count reaches
zero, the Debugger will take the breakpoint, stop
execution of the program, and reset the
breakpoint-countto 1.

Take breakpoint
and reset
breakpoint
count to 1

Figure 4-1. Breakpoint Logic Diagram

093-000195-02

4-1

Breakpoint-Related Commands

DataGeneral

SOFTWARE DOCUMENTATION

NOTE: You must specify breakpoints only at
executable instructions. Do not set
breakpoints at the following locations: data
words, instructions modified during program
execution, or the second word of a two-word
instruction.

Examples:

B447)

Set unconditional breakpoint at location 447.

B447;#0%EQ%36)

Set conditional breakpoint at location 447. (See
Chapter 2 for a description of these logical operators.)
Note that breakpoint count is omitted, so the Debugger
automatically sets the count to 1. When the breakpoint
is reached, the Debugger evaluates the
breakpoint-condition as follows:

• Compare the contents of accumulator against the
constant value 36.

• If the result is nonzero (true), take the breakpoint.

• If the evaluation result is zero (false), skip the
breakpoint and do not alter the breakpoint count.

B447;@1503%GT%0%OR%#3%NE%6)

Same type of conditional breakpoint as above. The
Debugger evaluates the breakpoint-condition as
follows: if the contents of location 1 503 are greater
than 0, or if accumulator 3 does not contain 6, take the
breakpoint.

B447;;6)

Set unconditional breakpoint. Note that if you omit a
breakpoint-condition and include a breakpoint count
you must include an additional semicolon (;). In this
example, the count is set to 6. Thus, the program will
pass through the breakpoint 6 times before the
Debugger takes the breakpoint; but the Debugger will
decrement the count each time the program passes
through it, until it equals zero. When the count equals
zero, the Debugger takes the breakpoint, then resets
the count to 1. In this way, you can set the breakpoint
count to skip breakpoints as required.

B447;#0%NE%0;6)

Set conditional breakpoint; the breakpoint-count is set
to six. Each time the Debugger evaluates the
breakpoint-condition and the result is nonzero (true),
it decrements the count by one. When the count equals
0, the Debugger takes the breakpoint.

Licensed Material - Property of Data General Corporation

447:001500 +
B)

The first command displays the contents of location
447. The following B command sets an unconditional
breakpoint at location 447.

NOTE: Breakpoint commands do not apply to the
Disk File Editor.

Examine Breakpoints

To help you examine breakpoints, the Debugger
assigns each one a number in the order in which you set
them. The Debugger will display this number when the
breakpoint is taken, or when you request the Debugger
to display the existing breakpoints (see below). The
Debugger returns the breakpoint identification number
as:

Inumber

Because Inumber is a symbol for the breakpoint's
location, you can use it interchangeably with the
location address in any valid Debugger expression.

Example:

Assume you are setting the first breakpoint:

1 400: 0340 J 6 + (Display contents of location 1 400. )

B) (Set unconditional breakpoint at

location 1400. Since this is the first
breakpoint, the Debugger assigns the
value !0 to this breakpoint.)

!0= 1400 (Display the value of the breakpoint

identification number.)

!0: 034016 + (Display the breakpoint location's
contents.)

To display a complete list of your program breakpoint
locations, key in the following command:

?B)

Example:

?B)

!0 7554 (where 7554 is the breakpoint address)

11 7553 6 (where 6 is the current breakpoint count

value)

4-2

093-000195-02

Licensed Material - Property of Data General Corporation

Delete Breakpoints

There are two different commands which you can use
to delete breakpoints. To delete one or more specific
breakpoints, key in the following command:

DB address [; address] ...)

where address is the previously-set breakpoint address.

Examples:

DB 7553 )

DB7554;7553)

DB !0;7553 )

Note that you can use the breakpoint identification

number when deleting breakpoints. To delete all

previously-set breakpoints, key in the following
command:

NOBRK)

Start User Program Execution

To start or proceed with execution of your program,
key in a Start User Program Execution command in
this format:

P [breakpoint-count] }

The Debugger starts your program execution from the
location contained in the program counter.

If you are at a breakpoint, you can change the
breakpoint-count by keying in breakpoint-count. It can
be any valid Debugger expression.

NOTE: "P" commands do not apply to the Disk File
Editor.

Examples:

P)

Starts or continues program execution with the
instruction whose address is in the program counter.
#P is a temporary variable maintained by the Debugger.
It represents the contents of the current program
counter. If you want to start program execution at a
point other than the one in the current program
counter, change the temporary variable representing
the current program counter (#P). Use the SET
command described later in this chapter. Temporary
variables are described under the SET command.

P3)

DataGeneral

SOFTWARE DOCUMENTATION

This assumes that your program is halted at a
breakpoint. The command resets the breakpoint count
to 3.

Set Variable Command

To change the value of an existing temporary variable,
or to create a new one, key in the following command:

SET variable-name; expression)

where expression is any valid Debugger expression,
and variable-name is the name of the variable you wish
to change or create. The Debugger will evaluate
expression and set the result as the value of
variable-name. If variable-name does not already exist,
the Debugger will create it, then assign it the value of
expression. Note that you also use this command to
modify the accumulators ( #0, #1 , #2, #3 ) the carry bit
( #C ), and the program execution address ( #P ).

Examples:

SET #0;-1

changes the value of accumulator to -1 .
SETFOO;62

changes the value of FOO to 62. If FOO does not exist,
the Debugger creates it and assigns it a value of 62.

Display Accumulator Command

To display the contents of accumulators through 3,
key in the following command:

?A)

Example:

?A)

# = 066666* 1 = 000001 # 2 = 000007 # 3 = 000011

Location-Related Commands

Display Contents of a Location

To display the contents of a location, key in the
following command:

[address]:

where address is any valid Debugger expression.

The Debugger displays the contents of the location
referenced by address in the current display modes

093-000195-02

4-3

Location-Related Commands

DataGeneral

SOFTWARE DOCUMENTATION

(display modes are explained in the next command). If
the address is omitted, the Debugger uses the data
contained in the last displayed word as the address, and
displays the contents of that location.

Examples:

31 05: 000050 +

displays contents of location 31 05.

.000007 +

uses the contents of the last word as the address and
displays its contents.

Additional examples of displaying a location's contents
are in Chapter 2.

Modify Contents of a Location

To modify the contents of a location, key one of the
following commands:

[address;] expression )
[address;] expression (CR/LF)
[address;] expression (SHIFT-N)

where address and expression can be any valid Debugger
expression.

If you terminate the command with new-line ), the
Debugger modifies the contents of the location and
displays the prompt ( + ) again. If you terminate the
command with (CR/LF), the Debugger modifies the
location's contents and displays the contents of the next
location; if you use |, the Debugger modifies the
location's contents, and displays the contents of the
previous location.

The Debugger evaluates expression and stores its
value in the location specified by address. If you omit
the address, it stores the expression's value in the
last-displayed location. In most cases you will want to
examine a location before you modify it.

• If address or the last-displayed location is a word
address, expression is stored as a word (16 bits).

• If it is a byte address, the data is stored as byte (8
bits).

• If it is a bit, the data is stored as a single bit.

If the expression evaluates to a value larger than that
which will fit in the addressed location (word, byte, or
bit), the Debugger gives you an error message.

Licensed Material - Property of Data General Corporation

Examples:

1400: 000011 + LDA3 16)

displays contents of location 1400; then stores the
instruction LDA 3 1 6 in location 1 400.

1503&Y1;1 06 (CR/LF)
1504:401

replaces the contents of the right byte in location 1 503
with 106; (CR/LF) causes the Debugger to display the
contents of location 1504 (401).

Display the Next Data Item

After the Debugger displays the contents of a location,
you can key in a (CR/LF) (ASCII 015) to display the
next data item. (The command increments the offset
part of the last-referenced item by 1 (word, byte, or
bit), and uses that address.)

Example:

Given the following addresses and their contents:

Address Contents

1 400 005602

1401 000003

1402 000050

1 403 000200

then,

MODEW) (changes address mode to

word addressing)

1400: 005602 + (CR/LF) (displays locations 1400, then

1401)

1401:000003 + (CR/LF) (displays location 1402)

1402:000050 +

MODEY) (changes address mode to

byte addressing)

1400 &Y: 013 + (CR/LF) (displays left byte, then right

byte in word 1400)
1 400 &Y 1:202 + (CR/LF) (displays left byte of location

1401).

1401 & Y:000 -4

We used &Y in the command 1400 &Y: because 1400
must be identified as a word, not a byte, location.

Display the Previous Data Item

When the Debugger displays the contents of a location,
you can type (SHIFT-N) ( j ) to display the previous
data item. (This command decrements the offset part
of the last-referenced item by 1 (word, byte, or bit),
and uses that address.)

4-4

093-000195-02

Licensed Material - Property of Data General Corporation

Example:

Given the same addresses and contents as in the
previous example, then,

MODE W) (changes address mode to word

addressing)
1 402: 000050+] (displays location 1402, then

1401)
1401:000003 + 1 (displays location 1400)

1400:005602 +

Display a Range of Data Items

To display a range of data items (words, bytes, or bits)
in your address space, key in the following command:

DISP address-1 ;address-2 [increment] [;condition] )

The Debugger searches through all data items
beginning at address-1 and ending with address-2. It
successively adds the increment to each address,
beginning with address-1, to select the locations it
analyzes. The size of each location (word, byte, or bit)
is specified by the address-mode character in
address-1, or by the default mode if you omit the
address-mode character. If you omit increment, then the
Debugger uses an increment of 1.

For each location it selects, the Debugger evaluates
condition (where condition is also any valid Debugger
expression). If the result is true (nonzero), then it
displays the item; otherwise the item is not displayed. If
condition does not appear in the command, the
Debugger unconditionally displays all items. Thus, you
can use condition in this command to search for specific
items in your address space.

Examples:

DISP 1400;1 500)

displays the contents of all word locations beginning at
address 1 400 through 1 500.

DISP 1400;1500;3)

displays the contents of every third word beginning at
address 1400 through 1500 (i.e., 1400, 1403, 1406,
etc.).

DISP 1400 &Y1;1 405 &W)

displays every byte beginning with the right byte in
location 1 400 and ending with the left byte in location
1405.

DataGeneral

SOFTWARE DOCUMENTATION

DISP 1400;1415;;@.%NE%0)

displays the contents of every location whose contents
are not equal to 0. Note that if you omit the increment
and include a condition, you must insert an additional
semicolon in place of the increment. The DOT symbol
references the location currently being analyzed by the
DISP command.

DISP 1 400 &Y1 ;1 430&W;2; @.%EQ% #3)

displays the contents of every other byte, starting with
the right byte of word 1 400 and ending with the left
byte in word 1 430, if the contents of the byte are equal
to the contents of accumulator 3.

Suppress Symbols

To suppress symbols defined within a range of
addresses, use the command:

NOSYM expressionl ;expression2)

where expression is any valid Debugger expression.

The Debugger will disregard any symbol defined
between expressionl and expression2. This will help
you avoid confusion caused by a large number of
symbols for the locations you are using.

Example:

Given the following addresses, symbols, and contents:

Address Symbol Contents

1 400 QAT 005602

1401 QAT+1 000003

1402 QAT + 2 000050

1 403 FUM 000200

1404 FUM -hi 000073

1405 FUM + 2 000652

1406 ZRK 003020

Without NOSYM, the dialog involved in displaying the
data would be:

^ 400: 005602 + (CR/LF)
QAT +1:000003+ (CR/LF)
QAT +2:000050+ (CR/LF)
FUM:000200 + (CR/LF)
FUM +1:000073 + (CR/LF)
FUM+2:000652+ (CR/LF)
ZRK:003020 +

093-000195-02

4-5

Location-Related Commands

DataGeneral

SOFTWARE DOCUMENTATION

The NOSYM command, however, will make the
Debugger ignore the symbols defined within the
specified range:

NOSYM QAT;ZRK)
1 400:005602+ (CR/LF)
Q AT +1:000003+ (CR/LF)
QAT+ 2:000050+ (CR/LF)
QAT+ 3:000200+ (CR/LF)
QAT+ 4:000073+ (CR/LF)
QAT+ 5:000652+ (CR/LF)
ZRK:003020 +

Display an ASCII String

To display an ASCII string, key in the following
command:

DSTR byte-address [;length] )

Byte-address is the address of the first byte of the
string to be displayed, and can be any valid Debugger
expression. If byte-address does not contain an
address-mode-character, the Debugger uses
byte-address as a byte offset. This allows you to
reference the contents of a location to get a byte
address for display.

The byte string must be terminated by a null. If the
byte string is greater than 131 characters, the Debugger
will display only 131 characters. If you want to display
fewer than 131 characters, specify length for the
maximum number of characters you want to display.

Examples:

Given the following addresses and their contents:

Address Contents

507

001220

(Byte address
of word 510)

510

040502

(AB)

511

041504

(CD)

512

042506

(EF)

513

000000

(NULL,NULL)

then,

DSTR 1220)
ABCDEF

DSTR @507;3)
ABC

DSTR510&Y 1)
BCDEF

Licensed Material - Property of Data General Corporation

Mode-Related Commands

The Debugger always displays the contents of locations
according to four types of display submodes: format,
radix, shift and sign. (This is in addition to the address
mode, which allows you to access the contents of
words, bytes, or bits.) The /om?<7f submode defines the
data representation (numeric, ASCII, symbolic, etc.);
the radix submode defines the base of the number
system used; the shift submode positions the data
within the word; and the sign submode specifies signed
or unsigned numbers. Once you set a display or address
submode, it will be maintained until you change it.

FORMAT SUBMODES 1

Format Submode
Character

Displays data as:

F

A 16-bit numeric constant.

H

An 8-bit numeric constant.

A

A pair of ASCII characters.

S

A symbol plus offset.

N

An instruction.

P

Single-precision floating-point data.

Q

Double-precision floating-point data.

When you enter the Debugger, the default format
submode is F (16-bit numeric constant).

-RADIX SUBMODES

Radix Submode

Action:

Character

B

Display data in binary format.

O

Display numeric constants in octal.

D

Display numeric constants in decimal
with a trailing period. The presence or
absence of a trailing period lets you
see at a glance whether numeric
constants are octal or decimal.

X

Display numeric constants in
hexadecimal (base 16).

When you enter the Debugger, the default radix
submode is O (octal).

4-6

093-000195-02

Licensed Material - Property of Data General Corporation

You can specify shifted data with the character:
SHIFT SUBMODE

DataGeneral

SOFTWARE DOCUMENTATION

Character

T expression

Action:

Display data shifted the number of
bits specified in expression.
Expression can be -15 to +15 bits.
If expression > 0, the data is shifted
left; if < 0, the data is shifted right.

Sign Submode

- SIGN SUBMODES .

Action:

Character

US

Display constants without regard
to sign (i.e., as 16-bit integers).

SI

Interpret the sign bit when
displaying constants. The
Debugger will display a negative
number with a minus sign

preceding it.

When you enter the Debugger, the default sign
submode is US (unsigned).

Address Submode
Character

ADDRESS MODES 1

Action:

W

Address offset references a 16-bit
word. When you display data in
this mode, you will display a 16-bit
word.

Y

Address offset references an 8-bit
byte. When you display data in this
mode, you will display an 8-bit
byte.

1

Address offset references a single
bit. When you display data in this
mode, you will display a single bit.

When you enter the Debugger, the default address
mode is W (word addressing).

To change the default format, radix, sign and address
mode characters, key in the following command:

MODE mode-character [;mode-character] ...)

(Of course, you can select only one mode character for
each of the four submode, and one address mode,
types, for a maximum of five mode-character
arguments per MODE command.)

Examples:

MODED:SI)

changes the radix and sign submodes.

MODESI;Y)

changes the sign submode and address mode.

MODEY;T-3;US;X;H)

changes all submodes and the address mode.

You can temporarily change the address mode for any
address by typing an ampersand (&) and the new
address mode character after the address. For example,
assume that W is the current address mode:

73: 020111+ 73&Y: 040 + 73&I: +

The first command displays the contents of location 73
in the default (word) mode; the second command
displays the contents of 73's left byte; the third
command displays the contents of 73's bit 0. Because
an offset address was omitted in the Y and I commands,
the Debugger assumed an offset of 0, and dealt only
with byte (left byte) and bit of location 73.

The default address mode remains W after these
commands.

Display Current Display Modes

To display the current display modes, key in the
following command:

?M)

Example:

?M)
WFOUS

Display Last Item with Different Display
Modes

To redisplay the last displayed item in different display
modes, enter one or more mode-changing characters in
the format:

mode-character [;mode-character] ... (ESC)

where ESC is the ESC key on your console.

Example:

1400:041520 + A(ESC) CP (displays 041520 as two

ASCII characters)

093-000195-02

4-7

Location-Related Commands

DataGeneral

SOFTWARE DOCUMENTATION

Floating-Point Accumulator-Related

Commands

Set Floating-Point Accumulator

To set a floating-point accumulator, key in the
following command:

SFP expression-1 ;expression-2)

where expression-1 is the number of the floating point
accumulator:

= floating-point accumulator
= floating-point accumulator 1
= floating-point accumulator 2
= floating-point accumulator 3

expression-2 is the value to be stored in the
accumulator.

Examples:

SFP0;29.6)

SFP 0;041 035 114631 114631 114631)

Both of these commands will store the same value in
floating-point accumulator 0. If you enter a series of
integers in the accumulator, the Debugger stores them
in four parts of the floating-point accumulator in the
order you entered them. If you enter fewer than four
integers, the remaining portions are set to zero. Thus,
if you are dealing with single precision numbers, enter
only two values. The Debugger will set the unused
parts of the floating-point accumulator to zero.

Display Contents of a Floating-Point
Accumulator

You can display the contents of one or all of the
floating-point accumulators by issuing the following
command:

IF [number])

where number is the number of the accumulator whose
contents are to be displayed. If number is omitted, the
contents of all floating-point accumulators are
displayed.

NOTE: This command does not apply to DEDIT.

You can examine the floating-point accumulators only
when the floating-point unit is in use; i.e., the current
task must have a floating-point save area defined.

Floating-Point Status

Words 1 and 2 of the floating-point status are accessible
by using the temporary symbols # FS1 and # FS2.

Licensed Material - Property of Data General Corporation

Display Linked Elements

If you have elements on a linked list, you can display
any or all of them by typing the following command:

LLIST address-of-1 st-element; [link-offset]
; [display-start-address]; [display-stop-address]
; [display-condition]; [terminator]
; [maximum-chain-length] )

where:

means:

address-of-1 st-element the address of the first
element in the linked list.

link-offset

display-start-address

display-stop-address

display-condition

offset into the element
containing the link. If you
omit this variable, it defaults
toO.

a Debugger address indicating
the location within the
element where the Debugger
is to start data display (NOTE:
this variable must be an offset,
not an octal address). The
Debugger adds this offset to
the address of the current list
element. The address mode
determines whether a word,
byte or bit is displayed. If you
omit this variable, the default
is the link offset.

a Debugger address indicating
the location within the
element where the Debugger
is to end data display. The
address mode of this address
will not change the data size
displayed. If you omit this
variable, the default is the
start-address; data is displayed
in single units automatically.

You can include a logical or
arithmetic expression to
govern the display operation.
It will be evaluated like any
other Debugger expression.
display-condition is analyzed
after the Debugger finds the
element, but before it displays
data. If the result of the
expression is (false) the
Debugger does not display the
element. It continues to
search through the chain. You
can use the DOT symbol (.) to

4-8

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

display-condition

terminator

maximum-chain-length

represent the address of the
element. For example, you
could examine status bits by
setting a condition such as:

(. + 1&I0)%EQ% 1

This test bit of word 1 of
each element in the linked list
before displaying it.

the terminator in the last link
word. If you omit this
variable, the Debugger will
treat a or -1 as the
terminator.

If you do not include a value
here, the default is 32
(decimal).

NOTE: If you omit a variable, you must still include
the semicolon that precedes it.

Compute an Expression and Display its
Result

To compute an expression and display its result, key in
the following command:

expression [;mode-character] ...=

where expression is any valid Debugger expression.

The Debugger evaluates the expression and displays its
value immediately following the equal sign. It also
stores the result in temporary variable # R.

Examples for computing an expression and displaying
its value are contained in Chapter 3.

You can evaluate an expression and display its result in
whatever mode you wish. To do so, key in the mode
characters. If mode characters are omitted, result is
displayed as a 16-bit octal constant.

Interpret an Error Code

The Debugger will display the text of an error message
if you key in the following command:

MES error-code)

where error-code is any valid Debugger expression.
The Debugger will display the text corresponding to
this code on your console.

Example:

Assume accumulator contains the value 25 in error
code:

MES #0}

FILE DOES NOT EXIST

Append a Symbol Table

The Debugger automatically uses your program's
symbol table when you are debugging. You can,
however, add additional symbols by entering the
following command:

STAB)

SYMBOL FILENAME? symbol-table-pathname

The Debugger then adds and uses the symbols in
symbol-table-pathname during the debugging session.

Example:

STAB)

SYMBOL FILENAME? XTRSYMS.ST)

This command includes the symbols in the file
XTRSYMS.ST as part of the debugging symbols.

LOG Commands

You can have all Debugger dialog saved in a file for
later examination by typing:

LOG)

to which the Debugger will reply:

FILENAME?

Type the name of the file you want the dialog to be sent
to from this point on. To close the file so that you can
save it for later use, or enqueue it for printing, type the
command:

CLOSE)

Debugging Shared Libraries

To debug a shared library, key in the following
command before you attempt to set a breakpoint.

SHARE)

FILENAME? library-name)

where library-name is an ASCII string naming the
library to be debugged. You must give this command or
you will not be able to set breakpoints in the library.

093-000195-02

4-9

Debugging Shared Libraries

DataGeneral

SOFTWARE DOCUMENTATION

Licensed Material - Property of Data General Corporation

You may set breakpoints in any part of the shared
library which your program uses, whether or not the
code is actually in your process area. After the code has
been loaded, you may examine and change locations
within it.

SHARE)

FILENAME 7FOO.SL)

Do not debug the original copy of a shared library. In
general, you should make copies of any shared program

before you debug it. If your program (or the system)
fails while you are debugging a shared library,
breakpoints may be left inserted throughout it.

Terminate the Debugger

You can terminate the Debugger and return to the CLI
by typing the command:

BYE)

End of Chapter

4-10

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Chapter 5
AOS Disk File Editor (DEDIT)

The Disk File Editor utility (DEDIT) allows you to
examine or modify locations in AOS disk files. You can
edit any kind of file on disk using DEDIT. For
example, after you have debugged a program, you
might want to change its program file on disk with
DEDIT to avoid reassembling and rebinding the source
program.

DEDIT uses a subset of Debugger commands. The
disk version of a program is not executing when you
run DEDIT; thus Debugger commands which control
breakpoints, program execution, special Debugger
variables, and accumulators are meaningless to
DEDIT. The only Debugger commands you cannot use
in the Disk File Editor are those that set, display, and
delete breakpoints, start user program execution and
set and display special Debugger variables. You can use
all the rest.

To execute the Disk File Editor (DEDIT), type the
following CLI command:

XEQ DEDIT [/switches] ...pathname)

pathname

The pathname of the file you want to
edit.

Optional Switches:

/I = pathname DEDIT (Debugger) commands will
come from the pathname specified.
This lets you build a file of Debugger
commands and apply it with a single
DEDIT command. The file should
contain commands in the normal
format (each command terminated by
new-line (or CR/LF) in some cases),
including the last command.

/S = pathname Include the symbol table file identified
by pathname (similar to the STAB
command).

/L = pathname Save all DEDIT commands in a log file
identified by pathname (similar to the
LOG command).

When DEDIT is running and has successfully loaded
the file you specified for editing, it displays the prompt:

+ (the same prompt as for the Debugger)

You type in commands immediately following the
prompt, using the Debugger commands and
expressions described in Chapters 2, 3, and 4 of this
manual (unless you have used the /I switch).

When you have finished editing with DEDIT, type

BYE)

to return to the CLI.

CAUTION: If you submit Disk File Editor commands
in an input file to DEDIT (instead of
issuing them one by one from a console):

• Do not insert any characters or spaces
between a colon command (:) and the
value to be inserted (the patch value).
For example:

1234:STA 0,X will not work
1234:STA 0,X will work

• You must end the DEDIT command
file with a BYE command.

End of Chapter

093-000195-02

5-1

AOS Disk File Editor (DEDIT)

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Appendix A
Error Messages

UNKNOWN COMMAND

This happens when you enter a command such as:
1234 instead of 1234:

The Debugger tries to interpret 1 234 as an instruction
instead of an address.

ILLEGAL ADDRESS

You have asked the Debugger to look at a location not
in your address base.

INSTRUCTION CANNOT HAVE BREAKPOINT

You are trying to put a breakpoint on an SCL
instruction or an XOP1. This is illegal.

DATA TOO LARGE FOR ADDRESSED LOCATION

You are trying to enter a larger number (or longer
string) than will fit in the addressed location.

ILLEGAL DISPLAY OR ADDRESS MODE

You used an undefined display or address mode
character, or you tried to shift and the expression for
the number of bits could not be computed.

NO MORE ROOM FOR SYMBOL TABLES

The Debugger can have no more than eight symbol
tables in use, including the DEBUG. ST, which is
always present.

UNKNOWN SYMBOL

You have used a symbol not defined in the symbol file.

ILLEGAL INSTRUCTION

You have tried to use as an instruction a symbol that is
not defined as an instruction.

NULL EXPRESSION

You didn't enter anything where an expression of
some kind is required.

EXPRESSION TOO LONG OR TOO COMPLEX

The expression is too long to fit in the Debugger's
work area.

UNKNOWN OPERATOR

You have used an operator other than + ,-,/, etc.

EXPRESSION SYNTAX OR UNKNOWN SYMBOL
ERROR

You have typed in something which the Debugger does
not know how to interpret, for example: %$AF!

COMMAND ACCEPTS NO ARGUMENTS

You have supplied arguments for a command which
does not use any.

ILLEGAL COMMAND FOR UTILITY

This happens when, for example, you try to insert a
breakpoint while using DEDIT.

LOG FILE ALREADY OPEN

You gave the LOG command while a previous LOG
command was still in operation.

NO MORE ROOM FOR NO-SYMBOL PAIRS

The Debugger will not accept more than eight NOSYM
commands.

ILLEGAL SYMBOL FILE

The file given as an argument to a STAB command is
not a symbol file.

ILLEGAL BREAKPOINT ADDRESS

You have referenced a nonexistent breakpoint.

093-000195-02

A-1

Error Messages

DaiaGeneral

SOFTWARE DOCUMENTATION

Licensed Material - Property of Data General Corporation

NO MORE SHARED LIBRARIES

You can debug no more than eight shared libraries at a
time. If you issue a SHARE command when you have
eight already, you will get this error.

TOO MANY SHARED SYMBOLS

You have tried to use more than one symbol from a
shared library in an expression. This is illegal.

SHARED SYMBOL NOT CURRENTLY MAPPED

You have tried to load at a location in a shared library
not in the address base.

FLOATING POINT ERROR

You are trying to set the floating-point accumulator,
and either the accumulator number is invalid, or one of
the numbers specified to go in one of the 16-bit parts of
an accumulator is too large.

LIBRARY NOT BEING DEBUGGED

You are trying to set a breakpoint in a library not being
shared.

TOO MANY CHARACTERS IN TEMPORARY VARIABLE
A temporary variable can have at most ten characters.

ROUTINE IN PROCESS OF LOAD

You are debugging a multitask program which calls
routines from a shared library, and the routine you
want to examine is being loaded at this time.

FLOATING POINT UNIT NOT IN USE

You are trying to use the floating-point unit without
having designated a floating-point share area for the
specific task.

End of Appendix

A-2

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Appendix B
DEBUG/DEDIT Command Formats

This appendix lists all breakpoint commands as a group,
then lists all other Debugger commands alphabetically.

Breakpoint Commands
(Do Not Apply to DEDIT)

SET A BREAKPOINT

B [address] [;breakpoint-condition][;breakpoint-count] }

DISPLAY EXISTING BREAKPOINTS

?B)

DELETE ONE OR MORE BREAKPOINTS

DB address [;address] ...)

DELETE ALL BREAKPOINTS

NOBRK)

All Other Commands

APPEND A SYMBOL TABLE

STAB)

FILENAME? symbol-table-name)

CHANGE DISPLAY AND ADDRESS MODE

MODE mode-character [;mode-character] ...)

CLOSE DIALOG LOG FILE

CLOSE)

COMPUTE EXPRESSION AND DISPLAY RESULT

expression [;mode-character] ...)

DEBUG A SHARED LIBRARY

SHARE)

FILENAME? library-name)

DISPLAY AN ASCII STRING

DSTR byte-address ['length])

DISPLAY CONTENTS OF ACCUMULATORS 0-3

(does not apply to DEDIT)
?A)

DISPLAY CONTENTS OF A FLOATING-POINT
ACCUMULATOR (does not apply to DEDIT)
IF [number] )

DISPLAY CONTENTS OF A LOCATION

[address]:

DISPLAY CURRENT DISPLAY MODES

?M)

DISPLAY LAST ITEM WITH DIFFERENT DISPLAY MODES

mode-character [mode-character] ...(ESC)

DISPLAY LINKED ELEMENTS

LLIST address-of-1 st-element; [link-offset]
; [display-start-address]; [display-stop-address]
; [display-condition]; [terminator]
; [maximum -chain-length]

DISPLAY NEXT DATA ITEM

(CR/LF)

DISPLAY PREVIOUS DATA ITEM

(SHIFT N)

DISPLAY A RANGE OF DATA ITEMS

DISP addressl ;address2 ['increment] '['condition] )

INTERPRET AN ERROR CODE

MES error-code)

MODIFY CONTENTS OF A LOCATION

[address;] expression ) or (CR/LF)

093-000195-02

B-1

All Other Commands

DataGeneral

SOFTWARE DOCUMENTATION

Licensed Material - Property of Data General Corporation

SAVE DIALOG IN A FILE

LOG)

FILENAME? log-filename)

SET FLOATING-POINT ACCUMULATOR (does not apply to

DEDIT)

SFP expressionl ;expression2)

SET THE VALUE OF A TEMPORARY VARIABLE

SET variablename;expression)

START USER PROGRAM EXECUTION (does not apply to
DEDIT) P [breakpoint-count] )

SUPPRESS NEW SYMBOLS

NOSYM expressionl ;expression2

TERMINATE DEBUGGING/EDITING

BYE)

End of Appendix

B-2

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Appendix C
Command Terminator Keys and ANSI and
Non-ANSI Standard Consoles

It is important to know whether your consoles are
ANSI or non-ANSI standard terminals, because this
determines which key you should press to enter a
new-line character (octal 012) and a (CR/LF) character
(octal 015) when using the Debugger or Disk File
Editor.

One way to tell is by checking the keyboard to see if it
has a new-line and a carriage-return key. This usually
means it is an ANSI standard terminal. If the keyboard
has a carriage-return key and a line-feed key, it is
usually a non-ANSI keyboard.

Here is an example of the three keys:

+ MODEW)

+ 1 400: 005602 + (CR/LF)

1401& W:000003 +

This user types MODE W, a mode command, and
terminates it with a new-line character. The user then
types 1400 and a colon, which displays the contents of
location 1400. The user then enters (CR/LF).
(CR/LF) instructs DEBUG/DEDIT to display the next
location and its contents.

Your terminal must be properly matched to its device
characteristics. You can test this by typing the CLI
command

CHARACTERISTICS)

from the console you want to test. The CLI will display
the mnemonics that represent the characteristics for
this console. If your terminal is ANSI standard, the
mnemonic NAS (Not ANSI Standard) should appear
on the line showing the characteristics that do not apply
(/OFF). If your terminal is non-ANSI standard, NAS
should appear on the line showing the characteristics
that do apply (/ON).

In the example below, the console is a Model 4010A
(hard-copy) terminal, which is non-ANSI standard.

) CHARACTERISTICS)
IHARDCOPY /LPP=66/CPL=80
IONISTINAS

IOFFISFFIEP1ISPO/RAFIRA TIOTT
IEOLIUCSIL T/ULC/PM/NRM/TO

If the NAS mnemonic appears on the wrong line for
your terminal, you can correct it

A. for this run of your CLI process, by typing the
command

CHARACTERISTICS/

(S?f} /na s)

B. or permanently, by having your system manager
regenerate the operating system to change the
device characteristics for this console.

When your console is properly matched to its
characteristics, on an ANSI terminal you press the
new-line key to input ASCII 012 (which we call the
new-line character, ) ) and you press the
carriage-return key to input ASCII 015 (which we call
(CR/LF)). On a non-ANSI terminal, you press the
carriage-return key, which AOS translates to new line,
ASCII 012. You press the line-feed key which AOS
translates to (CR/LF), ASCII 015.

You can use Table C-l to determine whether your
terminal is ANSI standard or not, if you know the
model number, and whether the NAS mnemonic
should be represented as OFF or ON in response to the
CHARACTERISTICS command.

093-000195-02

C-1

Command Terminator Keys and ANSI and
Non-ANSI Standard Consoles

DataGeneral

SOFTWARE DOCUMENTATION

Licensed Material - Property of Data General Corporation

Table C-1. New-Line and CR/LF Keys on ANSI and non-ANSI

Keyboards

Console Model Number

ANSI/
non-ANSI

NAS

If you press:

Result:

Called in AOS Manuals

401 OA (TTY)
6040 (TTY)
40101 (CRT)
6012 (CRT)

Non-ANSI

ON

RETURN

key

Translated
to 012

new-line

)

LINE
FEED

key

Translated
to 015

(CR/LF)

6052 and 6053 (CRT)
other

ANSI

OFF

NEW-
LINE
key

012

new-line

)

RETURN

key

015

(CR/LF)

End of Appendix

C-2

093-000195-02

Licensed Material - Property of Data General Corporation

DataGeneral

SOFTWARE DOCUMENTATION

Index

Within this index, the letter "f" following a page entry
means "and the following page".

+ (prompt) 1-1

= (keystroke command) 4-9

! (examine breakpoint) 4-2

: (keystroke command) 4-3

t or SHIFT-N (keystroke command) 4-4

A command 4-3
address modes 4-7
address-mode-characters 3-1,4-7

I (offset reference for single bit) 4-7

W (offset reference for 16-bit word) 4-7

Y (offset reference for 8-bit byte) 4-7
angle brackets 2-4
append a symbol table 4-9
%AND% 2-3

ANSI and non-ANSI terminal differences C-l
arithmetic operators 2-1

+ (addition and unary plus) 2-1

- (subtraction and unary minus) 2-1

* (multiplication) 2-1

/ (division) 2-1

(parenthesis for algebraic expressions) 2-1
ASCII characters

inserting 2-4

values for console keys C-lf
assembler instructions 2-4

B command 4-1
base address 3-1
Boolean operators 2-3

%AND% 2-3

%OR% 2-3

%XOR% 2-3
breakpoint commands 4-1 to 4-3

B (set breakpoint) 4-1

DB (delete a breakpoint) 4-3

! (examine breakpoint) 4-2

NOBRK (delete all breakpoints) 4-3
breakpoint condition 4-1
breakpoint-count 4-1

BYE command 1-1,4-10
byte addressing 4-7

change modes 4-7

CHARACTERISTICS command C-l

characters, inserting 2-4

CLOSE command 4-9

close log file 4-9

command-code 1-1

command summary B-lf

command types 1-1

comparing values 2-2

conditional breakpoints 4-1,4-2

consoles C-lf

CR/LF (keystroke command) 4-4

DB command 4-3
DEBUG command 1-1
Debugger command formats 1-1
DEDIT command 5-1
delete breakpoints 4-3
dialog, save Debugger 4-9
disk file editor (DEDIT) 5-lf
DISP command 4-5
display

accumulator 4-3

an ASCII string 4-6

floating-point accumulator 4-8

format 4-6

linked elements 4-8

location contents 4-3

modes command 4-7

next item 4-4

previous data item 4-4

radix 4-6

range of data items 4-5

result of an expression 4-9

shift 4-7

sign 4-7

submodes 4-6
dot symbol (.) 3-2
DSTR command 4-6

093-000195-02

lndex-1

Index

DataGeneral

SOFTWARE DOCUMENTATION

entering the Debugger 1-1

%EQ% 2-2

error messages 4-9, A-l

examine a breakpoint 4-2

expressions,

definition of 2-1

evaluation of 2-1, 2-3
expression operators,

arithmetic 2-1

Boolean 2-3

definition of 2-1

indirect (@) 2-2

logical 2-2

?F command 4-8
file editor, disk 5-1
floating-point status symbols 4-8

#FS1, #FS2 4-8
format submode characters

(F, H, A,S,N, P,Q) 4-6

%GE% 2-2
%GT% 2-2

I (address mode) 4-7
indirect operator (@) 2-2
instructions, Macroassembler 2-4
interpret an error code 4-9, A-l

keystroke commands 1-1,1-2
: 4-3

CR/LF 4-4
= 4-9
| (SHIFT-N) 4-4

linked elements, display 4-8
LLIST command 4-8
LOG command 4-9
log file

close 4-9

open 4-9
logical operators (for signed integers) 2-2
logical operators (for unsigned integers) 2-3
%LE% 2-2
%LT% 2-2

?M command 4-7
Macroassembler instructions 2-4
MES command 4-9
mode 4-6f

address (I, Y, W) 4-7

changing 4-7

command (MODE) 4-7

default 4-7

format (A, F, H, N, P, Q, S) 4-6

radix (B, D, O, X) 4-6

sign (SI, US) 4-7

Licensed Material - Property of Data General Corporation

NAS mnemonic C-lf
%NE% 2-2
new-line iv

NOBRK command 4-3
NOSYM command 4-5
notation conventions iv

offset (address) 3-1
open log file 4-9
operator

arithmetic 2-1

Boolean 2-3

evaluation order 2-3

indirect (@) 2-2

logical

signed 2-2
unsigned 2-3
%OR% 2-3

P command (start program) 4-3
prompt ( + ) 1-1

radix submode characters

(B,0,D,X) 4-6
redisplay with different display modes 4-7

save Debugger dialog 4-9

set breakpoint 4-1

SET command 1-2,4-3

set floating-point accumulator 4-8

SFP command 4-8

SHARE command 4-9

shared libraries 4-9

shift submode character (T) 4-7

sign submode characters (US, SI) 4-7

signed expressions 2-2

special symbols

(#P, #C, #0, #1 , #2, #3, ER, #FS1 , #FS2) 1-2
STAB command 4-9
start program (P command) 4-3
suppressing symbols 4-5

terminating the Debugger 1-1,4-10
temporary variables 1-2

%UGE% 2-3
%UGT% 2-3
%ULE% 2-3
%ULT% 2-3
unconditional breakpoints 4-1,4-2

W (address mode) 4-7

%XOR% 2-3

Y (address mode) 4-7

lndex-2

093-000195-02

Aw DataGeneral

Software Documentation Remarks Form

•C

How Do You Like This Manual?

Title_

No.

We wrote the book for you, and naturally we had to make certain assumptions about who you are and how you
would use it. Your comments will help us correct our assumptions and improve our manuals. Please take a few
minutes to respond.

If you have any comments on the software itself, please contact your Data General representative. If you wish to
order manuals, consult the Publications Catalog (01 2-330).

Who Are You?

How Do You Use This Manual?

□
□
□
□
□

What programming language(s) do you use?

EDP Manager

Senior System Analyst

Analyst/Programmer

Operator

Other

(List in order: 1 = Primary use)

Introduction to the product

Reference

Tutorial Text

Operating Guide

Do You Like The Manual?

Yes

Somewhat

No

□

□

□

Is the manual easy to read?

D

□

D

Is it easy to understand?

□

□

□

Is the topic order easy to follow?

D

□

□

Is the technical information accurate?

D

□

D

Can you easily find what you want?

□

□

U

Do the illustrations help you?

□

□

U

Does the manual tell you everything you need to know?

Comments?

(Please note page number and paragraph where applicable.)

From:

Name

.Title.

Address.

-Company.
-Date

SD-00742

FOLD DOWN

FIRST

FOLD DOWN

FIRST

CLASS

PERMIT

No. 26

Southboro

Mass. 01772

BUSINESS REPLY MAIL

No Postage Necessary if Mailed in the United States

Postage will be paid by:

Data General Corporation

Southboro, Massachusetts 01 772

ATTENTION: Software Documentation

FOLD UP

SECOND

FOLD UP

SD-00742'A

STAPLE